Instructions to use keras/vit_large_patch16_224_imagenet with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- KerasHub
How to use keras/vit_large_patch16_224_imagenet with KerasHub:
import keras_hub import keras # Load ImageClassifier model image_classifier = keras_hub.models.ImageClassifier.from_preset( "hf://keras/vit_large_patch16_224_imagenet", num_classes=2, ) # Fine-tune image_classifier.fit( x=keras.random.randint((32, 64, 64, 3), 0, 256), y=keras.random.randint((32, 1), 0, 2), ) # Classify image image_classifier.predict(keras.random.randint((1, 64, 64, 3), 0, 256))import keras_hub # Create a Backbone model unspecialized for any task backbone = keras_hub.models.Backbone.from_preset("hf://keras/vit_large_patch16_224_imagenet") - Keras
How to use keras/vit_large_patch16_224_imagenet with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://keras/vit_large_patch16_224_imagenet") - Notebooks
- Google Colab
- Kaggle
| { | |
| "module": "keras_hub.src.models.vit.vit_image_classifier", | |
| "class_name": "ViTImageClassifier", | |
| "config": { | |
| "backbone": { | |
| "module": "keras_hub.src.models.vit.vit_backbone", | |
| "class_name": "ViTBackbone", | |
| "config": { | |
| "name": "vi_t_backbone", | |
| "trainable": true, | |
| "dtype": { | |
| "module": "keras", | |
| "class_name": "DTypePolicy", | |
| "config": { | |
| "name": "float32" | |
| }, | |
| "registered_name": null | |
| }, | |
| "image_shape": [ | |
| 224, | |
| 224, | |
| 3 | |
| ], | |
| "patch_size": [ | |
| 16, | |
| 16 | |
| ], | |
| "num_layers": 24, | |
| "num_heads": 16, | |
| "hidden_dim": 1024, | |
| "mlp_dim": 4096, | |
| "dropout_rate": 0.0, | |
| "attention_dropout": 0.0, | |
| "layer_norm_epsilon": 1e-06, | |
| "use_mha_bias": true, | |
| "use_mlp_bias": true, | |
| "use_class_token": true, | |
| "use_patch_bias": true | |
| }, | |
| "registered_name": "keras_hub>ViTBackbone" | |
| }, | |
| "preprocessor": { | |
| "module": "keras_hub.src.models.vit.vit_image_classifier_preprocessor", | |
| "class_name": "ViTImageClassifierPreprocessor", | |
| "config": { | |
| "name": "vi_t_image_classifier_preprocessor", | |
| "trainable": true, | |
| "dtype": { | |
| "module": "keras", | |
| "class_name": "DTypePolicy", | |
| "config": { | |
| "name": "float32" | |
| }, | |
| "registered_name": null | |
| }, | |
| "image_converter": { | |
| "module": "keras_hub.src.models.vit.vit_image_converter", | |
| "class_name": "ViTImageConverter", | |
| "config": { | |
| "name": "vi_t_image_converter", | |
| "trainable": true, | |
| "dtype": { | |
| "module": "keras", | |
| "class_name": "DTypePolicy", | |
| "config": { | |
| "name": "float32" | |
| }, | |
| "registered_name": null | |
| }, | |
| "image_size": [ | |
| 224, | |
| 224 | |
| ], | |
| "scale": [ | |
| 0.00784313725490196, | |
| 0.00784313725490196, | |
| 0.00784313725490196 | |
| ], | |
| "offset": [ | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "interpolation": "bilinear", | |
| "antialias": false, | |
| "crop_to_aspect_ratio": true, | |
| "pad_to_aspect_ratio": false, | |
| "bounding_box_format": "yxyx" | |
| }, | |
| "registered_name": "keras_hub>ViTImageConverter" | |
| }, | |
| "config_file": "preprocessor.json" | |
| }, | |
| "registered_name": "keras_hub>ViTImageClassifierPreprocessor" | |
| }, | |
| "name": "vi_t_image_classifier", | |
| "num_classes": 1000, | |
| "pooling": "token", | |
| "activation": null, | |
| "dropout": 0.0, | |
| "intermediate_dim": null | |
| }, | |
| "registered_name": "keras_hub>ViTImageClassifier" | |
| } |